Intelligent Prediction of Multi-Factor-Oriented Ground Settlement During TBM Tunneling in Soft Soil

Tunneling-induced ground surface settlement is associated with many complex influencing factors. Beyond factors related to tunnel geometry and surrounding geological conditions, operational the shield machine are highly significant because of complexity shield-soil interactions. Distinguishing most relevant can be very difficult, for all seem affect tunneling-induced some degree, none clearly influential. In this research, a learning method adopted intelligently select features based on measured data form robust non-parametric model which make prediction. The recorded from real construction site were compiled 12 summarized. Using intelligent method, two other in addition cover depth–pitching angle rolling angle–were distinguished among feature candidates as those trough. Another new finding that advance rate does not emerge top 10 selected models observational used. generated was validated by comparing testing dataset performance dataset. Sensitivity analysis conducted evaluate contribution each factor. According results, engineers general practice should attend closely pitching during excavation soft soil conditions.